Search

CN-121977269-A - Dynamic optimizing group control system and method for central air conditioner

CN121977269ACN 121977269 ACN121977269 ACN 121977269ACN-121977269-A

Abstract

The application relates to the technical field of intelligent control of air conditioners and discloses a dynamic optimizing group control system and method of a central air conditioner. The method corresponds to the system. According to the application, the running parameters and the predicted cold load are collected in real time, a real-time optimization model aiming at minimizing the total power consumption of the system is periodically established and solved, so that the optimal target control parameters are obtained, and then the water chilling unit, the water pump and the cooling tower fan are subjected to cooperative frequency conversion and start-stop control, so that the state of the water system dynamically approaches the optimal target, and the continuous efficient running and energy saving of the central air conditioning system under the variable working conditions are realized.

Inventors

  • LI JINGEN

Assignees

  • 广州市赛科自动化控制设备有限公司

Dates

Publication Date
20260505
Application Date
20260203

Claims (10)

  1. 1. A dynamic optimizing group control system for a central air conditioner, comprising: The system comprises a data acquisition module, a control module and a control module, wherein the data acquisition module is configured to acquire system operation parameters in real time, and the system operation parameters comprise operation state parameters of a water chilling unit, a chilled water pump, a cooling water pump and a cooling tower fan, and temperature parameters, pressure parameters and flow parameters of a water system; The load prediction module is configured to calculate a system cold load predicted value of a future set time length through a prediction model based on historical load data, environmental parameters and an operation plan; The on-line optimization module is connected with the data acquisition module and the load prediction module and is configured to receive the cold load prediction value and the current system operation parameters in each optimization period, take the system total power consumption comprising the water chilling unit power consumption, the water pump power consumption and the cooling tower fan power consumption as targets, take the cold load prediction value as key constraint, establish a real-time optimization model comprising the water chilling unit number combination, the chilled water supply temperature, the water pump frequency and the cooling water supply temperature as optimization variables, solve the real-time optimization model and output a group of optimal target control parameters; and the control execution module is connected with the online optimization module and each controlled device and is configured to start and stop and frequency conversion control on the water chilling unit, the chilled water pump, the cooling water pump and the cooling tower fan according to the optimal target control parameters so as to enable the temperature, the pressure and the flow of the water system to approach to the target states corresponding to the optimal target control parameters.
  2. 2. The central air conditioner dynamic optimizing group control system according to claim 1, wherein the optimization cycle time length is dynamically shortened or lengthened based on the fluctuation severity of the cold load predicted value or the steady state of the system operation parameter.
  3. 3. The system according to claim 1, wherein the prediction model is a cyclic neural network model integrated with a time domain attention mechanism, and the model dynamically distributes historical load data weights through the attention mechanism to output the predicted value of the cooling load.
  4. 4. The dynamic optimizing group control system of a central air conditioner according to any one of claim 1, wherein the objective function of the real-time optimizing model is: Wherein, the For the total power consumption of the system, In order to operate the number of the water chilling units, Is the first The load rate of the machine set, 、 And Is the first to Real-time chilled water temperature of water chiller Temperature of cooling water The dynamic coefficient of performance of the device is related, For the rated power coefficient of the device, Is the operating frequency of the water pump or the fan.
  5. 5. The system of claim 4, wherein the process of solving the real-time optimization model comprises: Randomly initializing a particle population in a solution space formed by a water chilling unit number combination, a chilled water supply temperature set value and a water pump and fan frequency; Performing multi-generation iteration on the population, calculating the adaptability of each particle in each generation iteration, and updating the optimal positions of individuals and the population And a preset threshold value Dynamically calculating the variation probability of particles : , wherein, As the reference mutation probability, For adjusting the coefficient, and according to the variation probability Disturbing the particle position and updating the particle speed and position; And when the iteration termination condition is met, decoding the optimal position of the current generation population into the optimal target control parameter and outputting the optimal target control parameter.
  6. 6. The dynamic optimizing group control system of a central air conditioner of claim 1, wherein the on-line optimizing module is further configured to dynamically adjust constraints of the real-time optimizing model and re-trigger a solving process according to equipment failure signals, valve states or out-of-limit information of the system operating parameters in each optimizing period.
  7. 7. The dynamic optimizing group control system of a central air conditioner according to claims 1 to 6, further comprising a reference checking module, connected to the data acquisition module and the on-line optimizing module, configured to perform an on-line calibration process, comprising in sequence: acquiring current system operation parameters from the data acquisition module, and calculating theoretical energy efficiency coefficients according to the equipment performance parameters and the current operation boundary conditions used for establishing the real-time optimization model in the online optimization module ; Calculating actual operation energy efficiency coefficient based on current system operation parameters And analyzing the continuous deviation from the theoretical energy efficiency coefficient; If the continuous deviation exceeds the set threshold, determining that the model is mismatched, and correcting the equipment performance parameters in the online optimization module by the following formula Calculated by the following formula: Wherein, the In order for the rate of learning to be high, For theoretical energy efficiency versus performance parameter Generalized inverse of jacobian matrix.
  8. 8. The central air conditioner dynamic optimizing group control system according to any one of claims 1 to 6, wherein the control execution module includes a model predictive control unit configured to execute an instruction generation process that sequentially includes: acquiring a temperature target value and a pressure target value in the optimal target control parameters as a set track; Predicting the temperature and pressure parameters of the water system in a future time domain based on the current system state parameters at each control moment in a control period shorter than the optimization period, and constructing a local optimization problem by taking the minimized tracking error between the predicted output and the set track and the minimized variable quantity of the frequency command of the frequency converter in the adjacent control period as a local optimization objective function; based on the local optimization problem, the variable quantity of the frequency command of the frequency converter during the adjacent control period is solved, and the obtained first control increment corresponding to the current control moment is converted into a frequency adjustment command which is issued to the corresponding frequency converter for execution.
  9. 9. The dynamic optimizing group control system of a central air conditioner according to claim 8, wherein the expression of the local optimization objective function is: Wherein, the To predict the time domain; to control the time domain; To at the same time Time pair Predicted values of temperature parameters and pressure parameters of the water system at moment; Is that Setting a track of time; To at the same time Time pair A predicted value of the frequency command variation of the time frequency converter; To at the same time Time pair A predicted value of a frequency command of the frequency converter issued at the moment; setting the frequency command as a frequency value corresponding to the optimal target control parameter; 、 And As a matrix of weights, the weight matrix, and weight matrix The weight coefficient corresponding to each device is dynamically adjusted according to the accumulated running time of the device, the longer the running time is, the larger the weight is, the weight matrix The weight coefficient of the corresponding cooling water temperature tracking error is inversely proportional to the energy efficiency coefficient value of the water chilling unit calculated in real time according to the current system operation parameters.
  10. 10. A dynamic optimizing group control method for a central air conditioner is characterized by comprising the following steps: Collecting system operation parameters in real time, wherein the system operation parameters comprise operation state parameters of a water chilling unit, a chilled water pump, a cooling water pump and a cooling tower fan, and temperature parameters, pressure parameters and flow parameters of the water system; calculating a system cold load predicted value of a future set time length through a prediction model based on historical load data, environmental parameters and an operation plan; In each optimization period, receiving the cold load predicted value and the current system operation parameters, taking the total system power consumption including the water chilling unit power consumption, the water pump power consumption and the cooling tower fan power consumption as targets, taking the cold load predicted value as key constraint, and establishing a real-time optimization model including the water chilling unit number combination, the chilled water supply temperature, the water pump frequency and the cooling water supply temperature as optimization variables; And solving the real-time optimization model, outputting a group of optimal target control parameters, and performing start-stop and variable-frequency control on a water chilling unit, a chilled water pump, a cooling water pump and a cooling tower fan according to the optimal target control parameters so as to enable the temperature, the pressure and the flow of the water system to approach to target states corresponding to the optimal target control parameters.

Description

Dynamic optimizing group control system and method for central air conditioner Technical Field The application relates to the technical field of intelligent control of air conditioners, in particular to a dynamic optimizing group control system and method for a central air conditioner. Background A central air conditioning system in a large public building is one of the main energy consumption devices, and the operation energy efficiency of the central air conditioning system is important. In a conventional central air conditioner room group control system, the control strategy is mostly dependent on fixed operation parameter setting or local adjustment based on a single parameter (such as host loading adjustment only according to backwater temperature). However, the load demand of the air conditioning system and the external environment parameters are continuously and dynamically changed, and the energy consumption characteristics of the cold water host, the water pump, the cooling tower and other devices are mutually coupled. The prior art lacks a control mechanism capable of performing collaborative, dynamic and global optimization on all key controlled devices in an air conditioning system based on real-time and predictive data. The system is often operated in a non-optimal state, all the devices are mutually independent, the maximization of the overall energy efficiency of the system can not be realized, and unnecessary energy waste is caused. Disclosure of Invention Based on the above, it is necessary to provide a central air conditioner dynamic optimizing group control system and method for realizing dynamic global optimizing. In one aspect, the present application provides a central air conditioner dynamic optimizing group control system, comprising: The system comprises a data acquisition module, a control module and a control module, wherein the data acquisition module is configured to acquire system operation parameters in real time, and the system operation parameters comprise operation state parameters of a water chilling unit, a chilled water pump, a cooling water pump and a cooling tower fan, and temperature parameters, pressure parameters and flow parameters of a water system; The load prediction module is configured to calculate a system cold load predicted value of a future set time length through a prediction model based on historical load data, environmental parameters and an operation plan; The on-line optimization module is connected with the data acquisition module and the load prediction module and is configured to receive the cold load prediction value and the current system operation parameters in each optimization period, take the system total power consumption comprising the water chilling unit power consumption, the water pump power consumption and the cooling tower fan power consumption as targets, take the cold load prediction value as key constraint, establish a real-time optimization model comprising the water chilling unit number combination, the chilled water supply temperature, the water pump frequency and the cooling water supply temperature as optimization variables, solve the real-time optimization model and output a group of optimal target control parameters; and the control execution module is connected with the online optimization module and each controlled device and is configured to start and stop and frequency conversion control on the water chilling unit, the chilled water pump, the cooling water pump and the cooling tower fan according to the optimal target control parameters so as to enable the temperature, the pressure and the flow of the water system to approach to the target states corresponding to the optimal target control parameters. In one embodiment, the optimization cycle duration is dynamically shortened or lengthened based on the severity of the fluctuation of the predicted cold load value or the steady state of the system operating parameters. In one embodiment, the prediction model is a cyclic neural network model fused with a time domain attention mechanism, and the model dynamically distributes historical load data weights through the attention mechanism to output the cold load predicted value. In one embodiment, the objective function of the real-time optimization model is: Wherein, the For the total power consumption of the system,In order to operate the number of the water chilling units,Is the firstThe load rate of the machine set,、AndIs the first toReal-time chilled water temperature of water chillerTemperature of cooling waterThe dynamic coefficient of performance of the device is related,For the rated power coefficient of the device,Is the operating frequency of the water pump or the fan. In one embodiment, the solving process of the real-time optimization model includes: Randomly initializing a particle population in a solution space formed by a water chilling unit number combination, a chilled water supply temperature set value and a w